I can see the difficulty in making the recessed vertical notch within the mortise using only a chisel (assuming one has no router). Would having a slope (as near vertical as possible) and originating at the exterior of the post on one side or both, rather than a vertical recessed face prove easier to make and still provide an acceptably secure connection? I don't have SketchUp to illustrate, but I think you can visualize what I mean. It seems much of the ultimate security falls on the wedges anyway and the visual result depends on neat cutting regardless.
Is the recess to prevent a visible gap due to moisture movement or just to provide maximum torsional rigidity (the notch obviously to prevent the stretcher from pulling out of the post)?
Stretcher to Post Joinery
- Chris Hall
- Site Admin
- Contact:
- Location: Greenfield, Massachusetts
Post
You could eliminate the notch within the post altogether, and widen the stretcher notch out to be the same width as the post, like this:
The problem with this approach however is that you lose stretcher strength against vertical loads, as the underside of the stretcher is no longer supported as it was when housed. The stretcher is weakened by the notching more than if the joint were housed and the entire stretcher section were to be carried by the mortise housing. The same problem applies to the idea you mentioned of making the shallow notches in the post faces into sloped abutments instead, in the interest of doing the work simply with fewer chisels or non-specialized tools.
Further, by widening the notch like that, you make the joint a little more prone to loosening due to shrinkage in the post. Shrinkage is a percentage after all, so a wider notch on the stretcher spanning the post entirely means that shrinkage or swelling in the post will have a slightly larger (and obviously visible) effect than if you house the stretcher. The same can be said with the half dovetail version which is un-housed, as compared to the housed version. I mention this issue in regards to the mechanical loss that results from the shrinkage, not the aesthetic one of a visible gap. Again, these types of joints are not generally meant for use on furniture of other places where there might be a need for highly refined visible joinery. A sawhorse, is somewhere closer to the rough framing of architecture in the scheme of things, but then again, one might make a 'sawhorse' as a furniture piece as well, used to support a table top or similar.
Re: Stretcher to Post Joinery
I think it is worth noting that the stretcher connections discussed so far are primarily undertaken with building framing elements which are hidden in walls or under floors. So, aesthetics in that case takes a back seat to structural and ease-of-fabrication concerns.SteveM wrote:I can see the difficulty in making the recessed vertical notch within the mortise using only a chisel (assuming one has no router). Would having a slope (as near vertical as possible) and originating at the exterior of the post on one side or both, rather than a vertical recessed face prove easier to make and still provide an acceptably secure connection? I don't have SketchUp to illustrate, but I think you can visualize what I mean. It seems much of the ultimate security falls on the wedges anyway and the visual result depends on neat cutting regardless.
Is the recess to prevent a visible gap due to moisture movement or just to provide maximum torsional rigidity (the notch obviously to prevent the stretcher from pulling out of the post)?
You could eliminate the notch within the post altogether, and widen the stretcher notch out to be the same width as the post, like this:
The problem with this approach however is that you lose stretcher strength against vertical loads, as the underside of the stretcher is no longer supported as it was when housed. The stretcher is weakened by the notching more than if the joint were housed and the entire stretcher section were to be carried by the mortise housing. The same problem applies to the idea you mentioned of making the shallow notches in the post faces into sloped abutments instead, in the interest of doing the work simply with fewer chisels or non-specialized tools.
Further, by widening the notch like that, you make the joint a little more prone to loosening due to shrinkage in the post. Shrinkage is a percentage after all, so a wider notch on the stretcher spanning the post entirely means that shrinkage or swelling in the post will have a slightly larger (and obviously visible) effect than if you house the stretcher. The same can be said with the half dovetail version which is un-housed, as compared to the housed version. I mention this issue in regards to the mechanical loss that results from the shrinkage, not the aesthetic one of a visible gap. Again, these types of joints are not generally meant for use on furniture of other places where there might be a need for highly refined visible joinery. A sawhorse, is somewhere closer to the rough framing of architecture in the scheme of things, but then again, one might make a 'sawhorse' as a furniture piece as well, used to support a table top or similar.
- Brian
- Deshi
Post
Re: Stretcher to Post Joinery
Chris,
Would it make sense to set the slope of the abutment inward slightly so that as the joint is tightened the notch will compress the fibers of the abutment? Does it make sense if the adjoining surface on the inside of the notch is set to the same angle?
Would it make sense to set the slope of the abutment inward slightly so that as the joint is tightened the notch will compress the fibers of the abutment? Does it make sense if the adjoining surface on the inside of the notch is set to the same angle?
- Chris Hall
- Site Admin
- Contact:
- Location: Greenfield, Massachusetts
Post
The angled abutments could also work, but I'm not sure you gain a whole lot by that over the chamfered interference fit, and the processing of the angle on the side grain of the post notch would be trickier to execute, to measure and I suspect more problems would be encountered in trying to fit that together well. The innards of the joints are, after all, not visible when the joint is being drawn up tight, so determining the problem areas is less straightforward.
Re: Stretcher to Post Joinery
I think the more straightforward route would be to chamfer the points in the notch which contact one another as the stretcher is slid downwards onto the cog, and make the width of the notch on the stretcher a hair smaller than the one inside the post, thus gaining a slight interference fit, compressing the grain in the post as the stretcher is brought down. Obviously the amount of compression would be gauged relative to the compressibility of the particular wood you are using. On a softwood (where this joint is typically used), you could probably get away with +1mm fatness on the post notch, but if using, say, jatoba, the compression amount might be 0.1mm or so.Brian wrote:Chris,
Would it make sense to set the slope of the abutment inward slightly so that as the joint is tightened the notch will compress the fibers of the abutment? Does it make sense if the adjoining surface on the inside of the notch is set to the same angle?
The angled abutments could also work, but I'm not sure you gain a whole lot by that over the chamfered interference fit, and the processing of the angle on the side grain of the post notch would be trickier to execute, to measure and I suspect more problems would be encountered in trying to fit that together well. The innards of the joints are, after all, not visible when the joint is being drawn up tight, so determining the problem areas is less straightforward.
- Brian
- Deshi
Post
Re: Stretcher to Post Joinery
The chamfer makes perfect sense, that seems a bit easier to be exacting than with the angled abutments.
- Chris Hall
- Site Admin
- Contact:
- Location: Greenfield, Massachusetts
Post
Re: Stretcher to Post Joinery
One reason these joints with the stretcher being 1/3 the size of the post face, are better suited to hidden architectural work than furniture concerns the issue of chamfering the stretcher. If the pieces are part of rough framing and are unseen, then, other than cursory efforts, chamfering doesn't enter into the equation. But in furniture or interior casework a similar connection between parts of similar proportions to one another requires chamfering of all the parts.
Typically with a through-mortise as such, the stretcher will have to have stopped chamfers. So, for the convenience and improved appearance of continuous chamfers, the stretcher should be thicker than the portion which goes through the post - i.e, it is a form of tenoned connection. And if there is to be a tenoned connection in the interest of having ease of chamfering, then there are a lot of possibilities when it comes to connecting stretcher to post with mortise and tenon, a cogged joint and half dovetail being but two.
Typically with a through-mortise as such, the stretcher will have to have stopped chamfers. So, for the convenience and improved appearance of continuous chamfers, the stretcher should be thicker than the portion which goes through the post - i.e, it is a form of tenoned connection. And if there is to be a tenoned connection in the interest of having ease of chamfering, then there are a lot of possibilities when it comes to connecting stretcher to post with mortise and tenon, a cogged joint and half dovetail being but two.
- SteveM
- Raw Log Import
Post
Re: Stretcher to Post Joinery
Thankfully, someone invented nails and glue
Seriously, tho', the explanation makes sense.
![Very Happy :D](./images/smilies/icon_e_biggrin.gif)
Post
I am providing photos of one set of sticks for one set of two saw horses (both sets are at the same progress) to support beams. I had started them before this explaining different alternatives to connections was started. After the first method was explained I decided to go ahead with the direction I had started and squared up the ends of the mortices for the stretcher ( have not squared up ends of other mortices) but after reviewing the 渡り顎 Watari Ago join I saw that that would be a better option. The only issue I have with that is that the overall length of the stretcher material I have cut is just long enough to reach thru the leg post since it matches mortices I have in the top beam. I could move the leg posts closer together and replace the top beams or just use them as is and make the cog narrower to accommodate the thru end.
Critique welcome.
Anyway here are photos and I will be holding on the project while waiting to see other options.
Re: Stretcher to Post Joinery
Chris and all,Jack_Ervin. wrote:Chris,Chris Hall wrote:The first case to deal with is a single stretcher meeting the post and terminating at that location:
I'm definitely with the others as to what you have to develop for different methods of connection of a stretcher to a post. I believe we all have procedures to learn or have confirmed. At present I have my mortises chopped to width but not finished to length. I have not cut tenons yet and will hold off until all methods of connections are explained. As to the above connection using pins, I'm assuming they are drawn to the shoulder and down.
I think Gadge and my end results for saw horses are the same.
Jack
I am providing photos of one set of sticks for one set of two saw horses (both sets are at the same progress) to support beams. I had started them before this explaining different alternatives to connections was started. After the first method was explained I decided to go ahead with the direction I had started and squared up the ends of the mortices for the stretcher ( have not squared up ends of other mortices) but after reviewing the 渡り顎 Watari Ago join I saw that that would be a better option. The only issue I have with that is that the overall length of the stretcher material I have cut is just long enough to reach thru the leg post since it matches mortices I have in the top beam. I could move the leg posts closer together and replace the top beams or just use them as is and make the cog narrower to accommodate the thru end.
Critique welcome.
Anyway here are photos and I will be holding on the project while waiting to see other options.
- Chris Hall
- Site Admin
- Contact:
- Location: Greenfield, Massachusetts
Post
Re: Stretcher to Post Joinery
How far beyond the cog would the stretcher extend if you did the watari ago connection?
Post
My statement above of better option was pertaining to strength/stability. I'm still weighing the option of the half dovetail and use all parts as presently cut. The stretcher would be left shouldered.
I'm sure open to suggestions.
Re: Stretcher to Post Joinery
If I decide that route, I would extend it by the width of the cog.Chris Hall wrote:How far beyond the cog would the stretcher extend if you did the watari ago connection?
My statement above of better option was pertaining to strength/stability. I'm still weighing the option of the half dovetail and use all parts as presently cut. The stretcher would be left shouldered.
I'm sure open to suggestions.
Return to “General Discussion”
Who is online
Users browsing this forum: No registered users and 6 guests